Ok so having had a bit of a rant, I posted a question on the OCA student forum, did some internet searches and took my pictures again. Some really useful responses to my question on the OCA forum and clearly this topic has been discussed at length over the last few months. Unfortunately I seem unable to post to this forum at present, not sure if it is a general problem or something to do with my settings.
I took the pictures again because I wanted to make sure that there was some constant in each of the pictures, for example, previously I had tried to use the same shutter speed then found that when I tried to measure the exposure in the black areas, the camera recorded the aperture as being too low. This time I set the aperture at f 5.6 and adjusted the shutter speed each time. For the first picture, I had to use a shutter speed of 1/15 sec before the highlight clipping warning went off, although once in Elements, I could see that some highlights were still clipped in an area where the light source was reflected directly on to the paper. I also found that although I had used matrix metering, the dark area, in fact the rest of the picture, was too dark so needed some exposure adjustment. The exposure for the very bright part of the white paper came out at 1/125 sec and the darkest part at 2.5 sec. Spot metering was used for both of these measurements. I calculated the difference between the 2 ends to be 9 stops. According to a review of my camera, a Nikon D7000 in Imaging Resource website, the dynamic range for this camera is 10 stops with Active D Lighting switched off as it was for this experiment. http://www.imaging-resource.com/PRODS/D7000/D7000IMATEST.HTM
When I checked the ‘pixel value’ of the white area I found that the only bit that had a value of anything like 255 was the area immediately around the clipped area so this was the only area of the white that was properly exposed. I didn’t really understand the significance of this until I sampled the pixels of the very dark areas of the lens and found that the area where shadow clipping was highlighted registered no pixels at all!
So far so good but what does it actually tell me? For a start, this was not a brilliant example because the desk lamp was not really bright enough and the white area was proportionally too big to get an accurate reading of the whole scene without clipping the highlights in the very bright area just above the cord of the lens bag. In order to get an image which looks reasonably exposed, I have had to increase exposure across the board, recover detail to get rid of the clipped highlights, reduce the blacks to get some detail back into the lens bag and increase brightness slightly. Although I have been shooting in RAW for a little while, it is not until now that I have had to perform so much processing before I could use the image but I can now see the value in it. I’m thinking that the solution must be to decide which part of the image is most important, expose for that and then perform any necessary adjustments at the RAW processing stage. The next exercise will put this to the test but I need to do this one again in better lighting conditions, also take some pictures with the Active D Lighting switched on to test the difference that this makes.